Glue Tubes Keypoints Project Computer Vision Project

Object detection is a powerful tool in hands of computer vision engineer.

However, object detection models, like Yolo5, can just identify the coordinates of the bounding box of an object, but they can't provide any information about object's orientation in space.

Such information can be provided by coordinates of the object's keypoints.

I was interested to get to know how to trainin a keypoint detection model, Keypoint RCNN, on a custom dataset. For that, I needed to create a simple dataset with objects related to one class, where each object was annotated with 2 keypoints, which would be enough to identify its orientation in space.

Glue tube was the ideal candidate to become such object. So, I've made 134 photos and annotated them.

Roboflow doesn't provide keypoint annotation functionality, but this task can be easily done by annotating keypoints with small rectangles.

With help of a custom script, we can convert these annotations

into these annotations

The tutorial how to convert rectangles into keypoints is here: https://medium.com/@alexppppp/how-to-annotate-keypoints-using-roboflow-9bc2aa8915cd

Also, the tutorial how to train a custom keypoint detection model is here: https://medium.com/@alexppppp/how-to-train-a-custom-keypoint-detection-model-with-pytorch-d9af90e111da

                        @misc{
                            glue-tubes-keypoints-project_dataset,
                            title = { Glue Tubes Keypoints Project Dataset },
                            type = { Open Source Dataset },
                            author = { My Workspace },
                            howpublished = { \url{ https://universe.roboflow.com/my-workspace-qpstx/glue-tubes-keypoints-project } },
                            url = { https://universe.roboflow.com/my-workspace-qpstx/glue-tubes-keypoints-project },
                            journal = { Roboflow Universe },
                            publisher = { Roboflow },
                            year = { 2021 },
                            month = { nov },
                            note = { visited on 2024-12-22 },
                            }